As is known, the basic idea of reactive power compensation is that the reactive power component required by the load need not be provided by the power supplier, which would create a greater burden on the grid, but instead, this reactive power effectively fluctuates back and forth between the load and a reactive compensation element solely on the consumption side.
A disadvantage of known arrangements for reactive power compensation is that these arrangements are not, or are only insufficiently able to respond to changes in the reactive power requirement of the load. In other words the compensation of the reactive power is only designed for a steady operating state of the load, and therefore the compensation itself is static. In so doing, a cos 0 from 0.85 to 0.95 is generally desired.